The Internet of Things (IoT) and artificial intelligence (AI) are the fastest-growing technological approaches globally. With the rising urban population, the concept of a smart city isn't new. To effectively integrate IoT and AI into remote healthcare monitoring (RHM) systems within smart cities, we must have a comprehensive grasp of smart city frameworks. Our proposed model focuses on gathering data from an auricular therapy device, a neurostimulator that provides non-invasive stimulation to the outer ear. This device communicates via Bluetooth, allowing data exchange between the patient's and doctor's phones. After collecting the brain signal data, it's processed to eliminate noise and is normalized. This data is then classified using the adaptive fuzzy based Bayesian metasalp neural network (AFBBMNN) combined with levy flight secure offloading analysis in Software Defined Networking (SDN). The results prominently emphasize the need for enhanced healthcare provision. This information is then relayed to doctors via a cloud-SDN module that comprises a communication phase, cloud server, and cloud database where the signals are stored. The proposed method offers promising outcomes, emphasizing its viability as an efficient tool for early neurological disease detection and treatment within a smart city healthcare framework.